The Pacing Problem: What 110,013 Boston Finishers Reveal About Running Smarter on Patriot’s Day

Across all pace groups and all four years combined, the average Boston Marathon runner’s second half was meaningfully slower than their first. But the scale varies enormously by ability level — and the pattern is remarkably consistent year after year.

Even sub-3:00 runners — experienced, well-trained marathoners — averaged a 5.0% positive split across all four years, and only 12.7% managed to negative split. The median positive split in that group was 4.2%. Boston eats everyone’s second half. The question is how much.

The first half of the Boston Marathon is a seduction. From the start in Hopkinton, runners descend roughly 300 feet over the first 10 miles. The legs feel fresh, the quads aren’t yet taking a beating, and the downhill gradient creates a false sense of effort — you’re running faster than your perceived effort suggests. The crowds in Wellesley around mile 12 add another surge of adrenaline.

Then comes the bill. The Newton Hills arrive between miles 16 and 21, culminating in the infamous Heartbreak Hill at mile 20.5. By this point, runners who went out too aggressively have already depleted glycogen stores and damaged quadriceps from the relentless downhill pounding. The hills don’t create the problem — they expose it.

We grouped runners into four finish-time tiers and measured their average positive split. With over 110,000 data points, the pattern is unmistakable and statistically robust.

The relationship between ability level and positive split is nearly linear — but the rate of deterioration accelerates. Sub-3:00 runners averaged +5.0% across four years; 4:00+ runners averaged +18.2%. For a 4:15 finisher, that means the second half was roughly 23 minutes slower than the first.

In the sub-3:00 tier, over half of all runners keep their positive split under 5% — that’s solid pacing. By the 4:00+ tier, more than half of all runners have positive splits exceeding 15%. These are runners losing 20–40 minutes in the second half compared to their first.

Does holding back in the first half actually produce faster finish times? To answer this, we grouped runners within each pace tier by their split behavior.

You might assume the world’s best marathoners simply run Boston the way they run every marathon — just faster. But the data reveals something far more interesting. The elites don’t just pace Boston better — they pace it differently. And there’s a striking divide in how men and women approach the course.

Even the fastest men in the world positive split Boston. Across four years, the top 25 male finishers averaged a +5.4% positive split, and only 2% managed to negative split. The course defeats even world-class men.

Elite women, however, tell a completely different story. Their average positive split was just +3.5%, and 15.4% of them negative split the course — nearly eight times the rate of elite men. In 2024, the contrast was even more dramatic.

Granular 5K split data from individual elite profiles reveals their approach to the course. Below is the 2024 men’s champion alongside a top-3 women’s finisher whose near-perfect pacing is a masterclass in restraint.

The contrast is striking. The men’s champion went through the first 20K in roughly even 14:15–14:30 splits, then slowed by nearly two full minutes per 5K through the Newton Hills and beyond. His 30–35K was 1 minute 39 seconds slower than his opening 5K.

The top women’s finisher? Her eight 5K splits ranged from 18:17 to 18:29. Total variation: 12 seconds. She ran the Newton Hills at virtually the same effort as the opening downhill. That is extraordinary pacing discipline.

To see how pacing strategy plays out at the mid-pack level, here are the actual 5K splits from two runners with similar ambitions but radically different outcomes.

The even-pacer’s slowest 5K was just 50 seconds off their fastest. Their range: 20:25 to 21:15. The crash runner’s 30–35K was 8 minutes and 16 seconds slower than their opening 5K — a complete collapse from 6:51/mi to 9:31/mi pace. Yet through 15K, these two runners were separated by barely a minute. The decision to crash was sealed in those first deceptively comfortable miles.

Boston’s course difficulty compounds with heat. Our four-year dataset reveals how dramatically conditions affect pacing outcomes.

2023 stands out: rain and cool temperatures produced the lowest positive split percentages across every tier. Even 4:00+ runners averaged just +13.9%. The warm 2024 edition was nearly twice as damaging for slower runners — the 4:00+ tier averaged +23.3%. The lesson is unambiguous: on warm race days, your pacing strategy must adjust or you will pay for it.

Only about 10% of all Boston Marathon finishers manage to negative split the course. These runners are doing something fundamentally different. Across all tiers, the data points to a consistent pattern: their first-half time is deliberately slower than even-pace would suggest.

Yes, that means if you’re chasing a 3:30 finish, you should run through the half in about 1:48–1:50 — feeling like you’re leaving minutes on the table. At Boston, you are not leaving minutes on the table. You are saving them for the Newton Hills.

The data couldn’t be clearer. At Boston, patience isn’t just a virtue — it’s a performance strategy worth anywhere from 5 to 18 minutes depending on your pace group. The course is designed to reward the disciplined and punish the eager. Across 110,013 runners and four years of racing in every condition from rain to warmth, the pattern holds.

The hardest part of running Boston isn’t Heartbreak Hill. It’s the first 10 miles. It’s holding back when everything feels easy, when the course is pulling you downhill, when the crowd energy is electric. The runners who run their best Boston — from the elite women who negative split with metronomic precision to the mid-pack qualifiers who resist the opening downhill — are the ones who save their best for when it matters.

On Patriot’s Day, your goal is simple: arrive at the base of the Newton Hills feeling like you haven’t raced yet. If you can do that, you’ll run the marathon you came for.